ZSutherland said:
Mike Mearls, and many other designers and commentators aside, have complained frequently about what he describes as "Mother May I" abilities.
Sidenote: If a player has the power to fight orcs then he has voted for there to be orcs in the game. A decent GM will chuck in orcs. Likewise a GM can tell players "This is going to be an orc-fighting game, make the preperations you need". Doesn't always have to be so cloak and dagger.
To me, the social skills describe the opposite affect. I think I'll call them "Baby Please!" abilities, which is cry of the hearts of DMs who've just watched their role-playing scene, encounter, or potentially their story go down the drain due to one stinking skill check.
This can only happen if the player suddenly levels up or brings in a new character to gain a skill where it wasn't there before. Look at the character sheets before the game and go "So who's got what skills here?". Design the adventure for those skills to come in handy.
Then don't use the skills as a 100% substitute for roleplaying.
Gather information, on a successful check, lets the player determine who they have to talk to in order to get the extra information. Most adventures have an extra bit of information that makes it easier. Robbing the countess de la marches can be done without knowing that she's a vampire, but it's easier with that extra information. On a successful gather info check, the player's told "Gingolan, the bartender at the gathered grape is rumoured to know a lot of the countess' secrets". Then getting that info out of Gingolan requires roleplaying through the conversation.
Sense Motive, again, give the extra information. "Gingolan, though he claims not to know anything about the countess is looking scared.", gives them the extra information they need to go back into the conversation and get what they need.
Bluff, firstly a successful bluff skill can only make them believe what you say. It's still up to the player to decide what lie to tell. That lie may, or may not, be the one that would get them out of trouble. You may convince the thieves' guild thugs that you work for the local crime boss, but if they're actually underagent crown guard then your bluff skill won't matter. This means sense motive and bluff combine nicely together, sense motive gives you some information to use and an intelligent player can funnel that info into the right lie for their bluff skill.
Intimidate, is the skill to be more intimidating than you should be. Judi Dench probably couldn't beat me up, but she has the social skills to make that irrelevant. Being intimidating while a 10' tall ogre with a massive greatsword can be taken for granted. But a halfling chained to a wall about to be tortured, might be able to use the intimidate skill to convince the torturer to back down.
Diplomacy, needs to be remembered that it has its limits. A male channeler could try to charm a whitecloak with diplomacy, but that whitecloak probably isn't interested in being friends. I rule that diplomacy can only be used if the other person is willing to have a conversation with you, and that people of good alignment can still fight evil people that are charming whilst evil aligned people may have no objection to robbing people they like.
In the end, rules exist to serve the game, most notably in places where player knowledge will not suffice.
Since I'm not an actor, when I GM I may not perfectly convey the NPC's emotional state and motivations. Sense motive covers the gap between the players experience and the character's. It fills in the descriptions that I don't give, particularly if I've cut down a 2 hour dinner party's conversation to a brief paragraph.
Why not just have the players elect the best spokesmen among them to do that job?
As the best spokesman from our gaming group, I'd like to ask for someone else to have a turn. For once I'd like to not play the bard... (not that it matters since we already use social skills in our games and everyone else can't seem to open their mouths without starting a war).